A Prediction Market for Macro-Economic Variables

Macro-economic forecasts are used extensively in industry and government even though the historical accuracy and reliability is disputed. Prediction markets have proven to successfully forecast the outcome of elections, sport events and product sales. In this paper we provide a detailed analysis of forecasts generated from a new prediction market for economic derivatives. The proposed market design is specifically designed to forecast macro-economic variables and differs significantly from previous ones. It solves some of the known problems such as low liquidity and partition-dependence framing effects. By using finance methodology we firstly show that the market is reasonably liquid in order to continuously generate forecasts. Secondly the market forecasts performed well in comparison to the ´Bloomberg´- survey forecasts. Thirdly forecasts generated by the market fulfill the weak-form forecast efficiency implying that forecasts contained all publicly available information.